The application's aims are to test hypotheses that: (1) stress-avoidant early parenting (Permissive) will produce low stress tolerance, regressive behavior, and low social assertiveness at adolescence; (2) stressful early parenting (Authoritarian, Rejecting) will produce low self-esteem and delinquency in "vulnerable" children but social assertiveness in "hardy" children; and (3) stress-monitoring early parenting (Authoritative) will produce well-adjusted, competent children who cope well with stress. Additional aims related to adolescent drug use are to: (1) establish personal and socialization antecedents and consequences of adolescent drug use and abuse; (2) construct drug use categories for the purpose of establishing "breaking points" where functional decrement (e.g. amotivational syndrome) occurs. General objectives are: (1) to identify origins in family socialization practices of developmental competencies at 4 crucial developmental stages--4-5, 9-10, 14-15, and early adulthood; and (2) to provide an empirical basis for prevention and intervention in serious adolescent maladjustment including drug abuse, supportive of efforts of emerging concerned parent groups. Families (130-160 depending upon the specific panel analysis) are drawn from middle-class caucasian residents of the San Francisco/East Bay area. At each time period, independent assessments are made by psychologists of the children and their parents using intensive clinical and cognitive interviews, naturalistic and structured observations of family interaction in the home and laboratory and of the child at school, self-report data, and psychometric tests. A distinction is made between stressors and stress, and between eustress and distress, with the position taken that stressful crises offer growth-enhancing as well as growth-retarding opportunities. Theoretically-guided hierarchical multiple regression methods are used to test causal hypotheses. Multivariate analyses of variance are used to compare child types on hypothesized antecedents and correlates. Multiple regression methods for analyzing panel data are employed to determine whether those aspects of parenting behaviors and child behaviors which are hypothesized to be associated with functional decrement resulting from drug use are in fact causally related to one another over time periods.